CHINESE HISTORY AND WRITING ABOUT RELIGION(S):

REFLECTIONS AT A CROSSROADS
ROBERT FORD CAMPANY
I want to offer some brief reflections on some ways in which the modern
Western notion of religion shapes our writing of Chinese history and
some ways in which Chinese history might impact our use of the notion
of religion1. I will not discuss modern (by which I mean post-1900)
Chinese usages of the term zongjiao , which, as has long been
known, was a neologism adopted around that time from the Japanese
term shky created expressly to translate the generic term religion, for which both premodern Chinese and Japanese lacked direct
equivalents.2 (Usages of zongjiao are more or less precisely equivalent to
uses of religion in Western languages.) Rather, I will discuss early
medieval (roughly ca. 100600 C.E.) Chinese usages relevant to our
topic. Western and modern as used here to modify notion of religion are of course shorthand for a discourse that may have had Western origins but has now spread worldwide. For better or worse, wielders
of the discourse of religion are no longer just Western people: they
include Chinese, Indian, Japanese, Korean, African, and many other
people3. If religion is by origin a Western category then it is one that
has now been globalized certainly in all areas touched by Western
educational systems, languages, intellectual taxonomies, media, or missionary activity.
I will do four things below. First I will summarize the points I made in
an already-published article concerning the modern Western discourse
on religions in the plural and the application of this discourse to phenomena in early medieval China. Then, turning to the singular term religion in its generic sense, that is, religion as a realm of concern as
_____________
1

Translations from Chinese are my own unless otherwise indicated, even where previous translations are cited for the readers convenience.
2
A rather impassioned counterargument to this notion was presented during our symposium, but I did not find it at all convincing; it is possible to turn up a few instances of
the compound zongjiao in medieval Chinese texts, but the term in those settings is not at
all close to the modern, Western-origin, folk or scholarly senses of religion.
3
Chidester, David, Savage Systems: Colonialism and Comparative Religion in Southern Africa, Charlottesville: University of Virginia Press 1996.

ROBERT FORD CAMPANY

opposed to other realms of concern with which it sharply contrasts, I will

ask three sets of questions:
(1) What conceptual and terminological pitfalls arise when writing
about religion in premodern China?
(2) How does Chinese religious history invite us to rethink our models
and assumptions about religion? Are there ways in which it could or
should affect how we think and write about religion at the level of both
methodology and, more subtly, of descriptive language?
(3) Did the generic notion of religion uniquely arise in the West, or
are there analogues in (for example) early medieval China? If there are
no premodern Chinese analogues, why might that be the case?
1. Synopsis of On the very idea of religions (in the modern West and in
early medieval China)
In this section I summarize an article of mine published a few years ago
on recent Western discourse on plural religions and on apparent analogues in early medieval China (Campany 2003).
The article opens with the following series of points:

Discourse on religions is first of all a linguistic affair, whatever

else it is. We often focus on big concepts, but at the working end
of writing about religions much is decided at the more concrete
level of descriptive and interpretive language.
Language is metaphorical in character. Underlying even the
most apparently neutral statements about religions will therefore
be found powerful metaphoric structures that shape the questions we ask and the assumptions we make.
Discourse about religions is rooted in Western language communities and histories. Other cultures in other times may and often do lack closely equivalent ways of demarcating phenomena.
The helpfulness of the category religions is not necessarily to
be measured by the extent to which people in the culture we are
describing would have recognized it as one of their own. It is a
second-order concept, ours to define. But use of this category
without any regard to whether, for example, medieval Chinese
usages work differently may lead to misunderstandings and projections, examples of which are discussed below.
That a culture such as early medieval China lacked one-for-one
equivalents of the category religions does not mean that it

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

lacked words and concepts that are analogous ones that did
similar sorts of work in similar sorts of rhetorical and social
situations.
I then presented some common metaphors that structure modern Western
discourse on religions: Religions are entities; religions are living organisms; religions are containers of people, ideas, texts; religions are personified agents; religions are marketable commodities; religions are armies and their spread is warfare. Implied is a critique of these: while they
may be useful for certain rhetorical, historiographic, etc., purposes, such
expressions are strictly and literally false: religions are not in fact any of
these things, including entities in any literal sense. Turning next to early
medieval China, I found a different set of metaphors, but ones employed
to do similar work; in other words, I found discourses about certain topics that are clearly analogous to Western talk of religions, ways of
speaking that we find in contexts where a Western author would understand the topic to be plural religions. We find,4 for example: founder or
paragon synecdoche (e.g., what we would term Buddhism, and not
merely the Buddha, inescapably intended by the way the term fo ,
normally meaning Buddha, is used in a certain discursive context); a
discourse of various ways or paths (dao and various compounds);
and a language of plural laws, methods, or regulations (fa ). We
also find a small set of standard metaphors in which the relations among
various species of these genuses were characterized for example, root
and branch (ben mo ) and various binary or triadic classification
schemes (e.g. left/right, yin/yang, life/death).
Some conclusions I drew from this discussion of Chinese metaphors
for things contextually analogous to religions were as follows. We do
find in early medieval Chinese texts some tendency to linguistically reify
phenomena that correspond to things called religions in Western discourses. It is not as if nothing approaching this idea ever occurred to
people in China in the first six centuries of the common era. On the other
hand, the standard early medieval Chinese metaphors are different and
have different implications. The way in which daos, etc., are spoken of
does not imply that they are total, encompassing systems or conceptual
frameworks, unlike the frequent case in Western writing about religions.
_____________
4
One thing we find very little of in early medieval times is the later-ubiquitous X
jiao , literally teachings of or about X, where X is a noun usually denoting a religious founder, sage, teacher, or deity. In modern Chinese this becomes standard usage for
every instance of what we would term religion, and there are ample premodern usages
of this term, but in early medieval texts such usage was not yet common.

ROBERT FORD CAMPANY

Nor do the Chinese metaphors imply the high degree of holistic integration implied by their Western counterparts. (An incidental remark on
translation: We are generally much better served by literal translations
than by ones that unhesitatingly map Chinese usages onto Western ones
without noting any differences. Fodao , for example, is much better
rendered as the way of Buddha[s] than as Buddhism.)
Such reification of things corresponding to the Western religions
arose, as did similar Western usages, in a particular sort of social and
historical context. That context was one of religious plurality and difference, of comparison and contrast occasioned by the forced juxtaposition
and the jostling for prestige of practitioners with different priorities. The
texts employing these metaphors arose at the boundaries between one set
of teachings-practices and another. In every case they were written from
the point of view of someone who, even when favoring one side or the
other, deems it possible to weigh both on the same scale and consider
them both as members of a common genus. Such situations naturally
involved contestation. It was, then, in the presence of others that communities began naming and defining themselves and each other and it
was in the midst of plurality that authors began using nominal forms to
denote entities that would be called religions in Western languages. 5
Such early medieval Chinese discourses are analogous to the early modern comparative religion of the West, but with one big disanalogy: to
speak of a religion in Western discourse is to imply a strong sense in
which it is a religion as opposed to some other type of thing some
non-religious type of thing as well as to differentiate it from other species in the genus religion. This was not quite true in early medieval
China. There were, naturally genuses of daos and of fas, but these were
not sharply demarcated as realms of concern or power from the non-dao
or the non-fa. I return to this point in section 4 below.
I concluded the article with two proposed alternatives to speaking of
religions as things in the world that act and have volition like personified
agents, grow (etc.) like plants, spread (etc.) like conquering armies, and
so on. Another way of putting this is that I set the metaphors aside to ask
what the purported things we call religions really, non-metaphorically
speaking, are in an ontologically rather literal sense. They are, among
other things, (1) repertoires of cultural resources (citing Ann Swidlers
_____________
5
Moerman (1965, 1967) found that even something as apparently basic as peoples ethnic self-identifications varied by situation, interlocutor, and topic of discussion.
The same might be even more true of discourse about collective bodies.

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

excellent work on cultural repertoires6) made and used over time by

many agents, and (2) imagined communities (citing Benedict Andersons
work on this theme 7) stretching across time and space. Thinking of them
in these ways, rather than as agents, plants, armies, containers, and the
like will open up new avenues of research and will close others that are
based on disanalogies, misunderstandings, and faulty projections.
2. Common pitfalls in writing religion in early medieval China
What conceptual and terminological pitfalls arise when authors
whether they are American, European, Chinese, Japanese, or other
write about religion in early medieval China, a culture which, as we will
see below, (at least to a very great extent) lacked analogues to our generic sense of religion as a kind or area of concern distinct from other?
Here I simply want to point out a few common assumptions about religion assumptions built inexorably into our very vocabulary for talking
about and imagining religion as a subject of discourse that create misunderstanding and falsify our representations of the Chinese religious
situations, often profoundly. Many of the difficulties faced by American
undergraduate students in understanding the subjects taught by people
like me stem from the disjunction between these assumptions and the
realia of Chinese religious history.
It is possible, in fact, to be much more specific. As many scholars
have noted, Christianity has often served as the implicit model of religion and religions generally, the case par excellence. (Religion means
things like Christianity8) All of the assumptions about religion that I
am about to point out result from the prevalence of the Christianity
model in discourse on religion and religions in general.
It is often assumed, for example, that a dichotomy between the natural
and the supernatural must be a feature of religion generally, so much so
that supernatural almost becomes synonymous with religious (when
describing an event) or deity (when describing a being). An early medieval Chinese genre of writing about human encounters with ghosts,
nature spirits, celestial gods, animal sprites, near-death experiences, and
other religiously tinged anomalies will thus end up being characterized
_____________
6
Swidler, Ann, Talk of Love: How Culture Matters, Chicago: University of Chicago
Press 2001.
7
Anderson, Benedict, Imagined Communities: Reflections on the Origin and Spread
of Nationalism, 2nd ed. London: Versa 1991.
8
Griffiths, Paul J., The Very Idea of Religion, in: First Things 103 (2000), 32.

ROBERT FORD CAMPANY

as supernatural fiction9 or as tales of the supernatural and the fantastic10 when, in fact, not only were these writings never intended as fiction in the standard sense of that term11 but the modifier supernatural
does not fit the subject matter.12 The lack of fit is not trivial, either. To
put it simply, and without lingering over the details of early medieval
Chinese notions of spirits, gods, the ordinary dead, and demons, many of
these beings were not understood as supernatural: they were neither
non-natural, nor were they metaphysically above nature, nor were they
non-material. They were natural in the sense that they were located in
this natural world that surrounds us and in which we, too, live; they were
material in that they (with the possible exception of a few deities at the
very top of the pantheon, so rarefied as to remain non-personified) had
bodies made of the same stuff qi that composed all other existent
phenomena. The overlay of the term supernatural onto such phenomena is understandable as a piece of economizing English-language shorthand but it creates a blind alley if the goal is to represent accurately in
English the phenomena being described.13 The assumption that such
phenomena and beings in China are correctly described as supernatural
is precisely one of the assumptions that instructors must disabuse Anglophone students of. Early medieval Chinese worldviews (like many other
worldviews in predominantly animistic, non-monotheistic civilizations)
tended not to dichotomize reality along the lines of nature/supernature.
Another cluster of assumptions basic to discourse on religion but misleading when it comes to describing early medieval China centers on
fideism. Adopting Western usages, many modern scholars working on
early medieval China write as if individuals beliefs were the salient
or even the only thing linking them to this or that religion, and as if a
list-like creed were the heart of any religion. To describe Daoism, then,
is to list the beliefs that were essential to it in all its forms and to which
_____________
9

Zhao, Xiaohuan, Classical Chinese Supernatural Fiction: A Morphological History,

Lewiston: Edwin Mellen Press 2005.
10
Kao, Karl S.Y. (ed.), Classical Chinese Tales of the Supernatural and the Fantastic,
Bloomington: Indiana University Press 1985.
11
Campany, Robert Ford, Strange Writing: Anomaly Accounts in Early Medieval
China, Albany: SUNY Press 1996.
12
The same usage applied to the same sorts of texts is seen in Kenneth DeWoskins
translation of the title of a famous work of this sort, Soushen ji , as In Search of
the Supernatural.
13
It might also be possible to construct a stipulative definition of supernatural that
would fit the early medieval Chinese religious scene, but this is rarely done and the modifier is simply applied without further discussion.

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

all Daoists assented; to describe a person as a Daoist is to signal that

he or she mentally assented to these beliefs. A concomitant research
program is then entailed: we sift through this individuals writings or
historical statements about his words or deeds, expecting to find (at least
in those records that happen to have survived) the clear signs of his inner
beliefs concerning ultimate things. A closely related tendency is to assume a denominational model of religious belonging or identification,
so that it is seen as making sense to ask, of a known individual, whether
he was a Daoist or something else. The unspoken assumption is that
he must have been only one at any given time an assumption again
based on a denominational and creed-based model of religious membership.14 It seems never to have occurred to these writers to ask whether
this entire complex of assumptions about religion, religions, and how
people relate to them a complex obviously based on post-sixteenthcentury Christianity is even applicable when describing a culture such
as early medieval China.15
An article was recently published, for example, in which the author
tried to make out whether a noted figure, Tao Qian (365427), was
really a Daoist, a Confucian, or perhaps even a Buddhist16. First,
then, it has been assumed that this very question makes sense in Tao
Qians time and place, that there is no better (i.e. more empirically accurate) way of imagining and talking about an individuals relationship to
a religion than this essentializing use of the copula. Next, in what does
this Daoistness etc. consist? In belief, of course. And how are Taos
beliefs to be ascertained, especially since he left writings that seem to
indicate positive interest in more than one ism? To answer this question the author takes the extraordinary approach of counting the number
of times Tao quotes from or alludes to works identified with one or the
other tradition, even referring at one point, without apparent irony, to
the quotation scoreboard17. Whichever traditions works Tao cited
_____________
14

Or, perhaps, a membership-based mode of religiosity, as if there were no other kind.

I would argue that it was perhaps at least partially applicable only when describing
two phenomena: the earliest generations of the Celestial Master movement and the taking
of Buddhist lay and monastic vows; but the matter is complex and an adequate discussion
is impossible here.
16
Holzman, Donald, A Dialogue with the Ancients: Tao Qians Interrogation of Confucius, in: S. Pearce / A. Spiro / P. Ebrey (eds.), Culture and Power in the
Reconstitution of the Chinese Realm, 200600, Cambridge, Mass.: Harvard University
Press 2001, 7598.
17
Holzman 2001, 85.
15

ROBERT FORD CAMPANY

most (in his works that happened to have survived the vicissitudes of
time) must be the winner!18
The often-unconscious assumptions about religion(s) made by writers
about a culture such as that of early medieval China profoundly shape
the questions asked and the research agendas posed. When the lack of fit
between those assumptions and the historical phenomena is as severe as
it is in the example just discussed, the result is problematic scholarship.
3. Ways in which religion in early medieval China invites us to rethink
our discourse on religion and religions
Here I want to ask what an understanding of the early medieval Chinese
religious scene might contribute to the modern, comparative study of
religion that is, to the very terms in which it is carried out. Whether or
not we find indigenous analogues to the sort of theorizing we are inclined to do ourselves,19 are there at least ways in which the Chinese
phenomena encourage us to rethink our assumptions? Are there even
perhaps models that we might appropriate into the discourse on religion?20 Might it be the case that aspects of early medieval Chinese religious life can teach us new ways of thinking about religion not just in
early medieval China but comparatively, at the level of categories? I will
briefly sketch five areas in which the answer might be yes. Further
development of these suggestions must await another occasion.
As has already been proposed (Buswell and Gimello 1992:117; cf.
Campany 2003:305), the fundamental Buddhist metaphor of path
_____________
18
For further discussion, and a suggestion of alternative models, see Campany, Robert
Ford, On the Very Idea of Religions (in the Modern West and in Early Medieval China),
in: History of Religions 42 (2003), 287319, and idem, Two Religious Thinkers of the
Early Eastern Jin: Gan Bao and Ge Hong in Multiple Contexts, in: Asia Major 3rd ser. 18
(2005), 175224.
19
Early on I tried to argue for such indigenous analogues (Campany, Robert Ford,
Survivals as Interpretive Strategy: A Sino-Western Comparative Case Study, in:
Method and Theory in the Study of Religion 2 (1990), 126; idem, Xunzi and Durkheim
as Theorists of Ritual Practice, in: F. Reynolds / D. Tracy (eds.), Discourse and
Practice, Albany: SUNY Press 1992, 197231.); now I am less confident that the analogies are close or true that the very same sort of thing was being done in both sets of
cases (Xunzi and Durkheim, early medieval Chinese folklore collectors and Tylor and
Frazer). To me at least, the question remains open.
20
The problem is really that of one-sidedness, a failure of mutuality. Where, for example, are the Hindu categories used to illumine Christianity, the Taoist concepts employed in analyzing Judaism, the shamanic themes applied to Islam? (Buswell and Gimello 1992, 1).

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

(Sanskrit mrga, Chinese dao ) suggests a way of envisioning and

describing individuals relations to religions that is an alternative to the
discourse of believing in this or that and belonging to X or simply
being an X-ist. One advantage is that the language of path (and its
attendant metaphors) takes belief off center stage and focuses instead
on movement in a certain direction along a series of prescribed, traditionsanctioned stations or stages, by means of further tradition-sanctioned
methods or practices. It understands participating in a religion, or being
religious, as movement toward a goal (or progression through a series of
goals) by means of things done, rather than as a simple toggle-switchlike assent to doctrinal propositions. Adopting this model in describing
not simply Buddhists but others as well would mean that the status of
propositions about reality and the role of individuals assent to them
would be reopened for investigation, rather than assumed; but it would
also mean that our attention would be broadened beyond beliefs to practices, and to how and why those practices were held to advance practitioners toward the specified goal(s).
A disadvantage of path-based models for religious participation is
that, like confessional models based on Christian denominationalism,
they are exclusivist: if it is a matter of paths, then by the very logic of the
metaphor one can only travel (or backtrack on, or try to leave, etc.) one
path at a time, just as, if it is a matter of the affirmation of beliefs, one is
presumed to assent to only one churchs set of creeds at a time. These are
disadvantages precisely because, as the example of Tao Qian above
shows, we find individuals not only in early medieval China! quite
often mixing, borrowing, combining, or playing between two or more
religions assertions or idioms, not just leaving one to adopt another but
at the same time actively using both at once. Such was the rule, not the
exception. So even the path metaphor has its limits when it comes to
providing a model that matches the suppleness and complexity of what
people often actually do. (People have a funny way of not conforming to
our models of how they should behave.) That is why I have elsewhere
suggested a model of repertoires of resources as more empirically accurate for describing how people often relate to religions and how they use
elements of their culture.21 Still, the path metaphor might be more useful
than the language of belief, belonging to, and being an X-ist when
it comes at least to describing the normative programs for spiritual ad_____________
21

Campany 2003, 317319; Campany 2005, 176177; based on Swidler 2001.

10

ROBERT FORD CAMPANY

vancement, salvation, or perfection taught by religious communities and

advocated in their texts.
Second, when we study how people actually related to religious institutions, ideas, groups, and activities in an environment such as early
medieval China, we see a much richer array of possibilities than confessional assent or denominational membership possibilities that also,
unlike the confessional/denominational model, include not just an on/off
toggle switch (where person P either believes in X-ism or he doesnt,
is an X-ist or isnt) but a subtler range of scenarios. We see people
acting as patrons or sponsors, for example, of advanced, specialist, virtuoso practitioners of esoteric arts: theirs was a different mode of participation than that of the ones on the path, but it was a mode that was recognized and respected as such and that brought certain rewards in its
own right. We also see people sponsoring the making of devotional objects or structures by donating resources, and again, this was a recognized, even formalized role, with a developed terminology and the careful recording of donors names for posterity right onto the objects or
structures they sponsored. We see individuals commissioning the making
of scripture copies or images so that the merit gained from this activity
could be transferred to their ancestors. Most ubiquitously, but for that
very reason often forgotten, we see individuals constituting the audiences before whom virtuoso practitioners carried out their esoteric or
ascetic disciplines and, often enough, performed their wonders and acted
out their strange violations of conventional mores: in short, to be a part
of the witnessing audience of the performance of a religious adept was
itself to play a sort of religious role. It was often only other peoples
recognition that accounted for his an individuals being deemed a transcendent, a wonder-working monk, etc., and being remembered as such in
the written record.22 Of course, any good historian of religions anywhere
including European Christianity in any period knows that there was a
rich array of roles and modes of participation available to people. But in
much writing on religion this richness is forgotten, and the language used
is by no means adequate to the complex realities found in the historical
record; often a person is said to simply be a this rather than a that, as if
there were nothing else to say on the matter.
Third, the history of religions in a place and time such as early medieval China suggests different models for understanding religious commu_____________
22

Campany, Robert Ford, Making Transcendents: Ascetics and Social Memory in

Early Medieval China, Honolulu: University of Hawaii Press 2009.

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

11

nities themselves, as well as pointing up the need for such models.

Rather than seeing them as congregations or denominations, we might, in
any cases, do better to see them as two things: repertoires of resources
on which individuals drew in shaping ideas, framing texts, and creating
rituals and regulations; and lineages of transmission through time and
across space, such transmission often formalized in rituals that were
among the most important in society at the time. That lineages were
sometimes fictively and retrospectively constructed does not lessen their
importance as a key way in which people in early medieval China
thought about and constituted religious communities. Scholars writing in
Western languages, particularly in English and recently, often speak of
religious traditions, perhaps because this seems a touch less reifying
and essentializing than religions, but in such discourse the etymological force of traditions is often ignored. To take it seriously would mean
seeing religions as, in part, imagined communities always in the process of being (re)constructed by linguistic and ritual acts of collective
memory and the transmission of ideas, values, techniques, and other
resources.
Fourth, as already hinted above, studying religion in early medieval
China suggests the need for a model of religiosity an understanding of
what it means to be religious that is grounded more in practice than
in belief per se. Affirmation of lists of beliefs was rarely, in short, the
defining marker or shaper of religious identity. If anything did perform
this function it was either the performance of certain practices or behaviors, especially ones that conspicuously marked a person as inhabiting a
role different from that of ordinary, default persons (e.g. shaving the
head, wearing a distinctive robe, abstaining from certain foods or from
eating at certain times of day, not marrying), or the formal taking of
vows to obey certain precepts that similarly marked the vow-taker as
different from others. Belief in certain beings, along with trust in ones
spiritual superiors and in the teachings of the tradition one was thereby
joining, were certainly not irrelevant,23 but the sheer fact of belief in this
rather than that was not the centrally defining feature.
Fifth, a few writers in late classical and early medieval China developed reflection on some religious topics that approach the sophistication
of theories that perhaps bear comparison with modern Western ana_____________
23
Case in point: T 2102, 52.82c2627, To accept this [the Buddhist teaching of the
far-distant results of karma over countless lifetimes] and to be able to believe is that not
difficult? ; cf. Zrcher 1971, 235.

12

ROBERT FORD CAMPANY

logues or at least contain ideas worth considering and applying in crosscultural analysis. Without elaborating further on these, I will simply
mention here two such areas: ritual (what it means, how it functions, its
effect on performers, how its symbolic code works and why symbolic
expression is used, etc.) and the path (Buddhist writers in particular developed a supple discourse that approaches being a path-ology).
4. Early medieval Chinese analogues to generic religion?
One Western author recently observed, summing up (and quoting) an
argument made late in his career by Wilfred Cantwell Smith: The modern notion of religion is au fond secularist; it sets up a dichotomy
between religion and what is not religion24. Another wrote: Christianity was rarely, if ever, thought of by Christians as one religion among
many: the idea that there is a genus called religion of which there are
many species did not gain much currency until the seventeenth century.
It is, by and large, a modern invention25. There were certainly many
things going on in early medieval China that anyone familiar with the
folk senses of the term would now say were religious in nature, and
there were phenomena that many would now say constituted religions
and to this latter notion there were even, as seen above, at least partial
indigenous analogues. But were there any analogues in early medieval
China to the putatively uniquely modern, Western notion of religion
(or its adjective (the) religious as opposed to (the) non-religious or
the secular) in the generic sense? I will argue here all too briefly and
sketchily that the answer is almost, but not quite. I will also suggest a
reason why such a notion, although briefly approximated in China, failed
to persist there.
So far as we know, texts, images, and monks representing the Buddhist religion were first introduced into China by the middle of the first
century C.E. By the beginning of the fifth century there were beginning
to be significant numbers of monks, nuns, temples, and monastic landholdings. Monasteries owned property, taking it out of the tax base and
redistributing production based on it. Buddhist devotion also soaked up
lay funds for donations and metals for the fashioning of images of
Buddhas and bodhisattvas. Furthermore, monks and nuns absented them_____________
24
Schssler Fiorenza, Francis, Religion: A Contested Site in Theology and the Study
of Religion, in: Harvard Theological Review 93 (2000), 17.
25
Griffiths 2000, 3132.

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

13

selves from some of the most basic Chinese social and ideological
frameworks: they left their families rather than creating descendants and
venerating ancestors through sacrificial ritual; they took the Buddhas
surname and renounced their own; they wore distinctive robes that, in the
case of monks, left one shoulder bare (which offended the sensibilities of
some Chinese), and shaved their heads the latter gesture seen as another renunciation of the family and of the veneration of ancestors, since,
ideologically speaking, ones body was inherited from the family line
and one was to carefully preserve it. Politically, too, the Buddhist sangha
posed a problem insofar as it set itself up as an authority independent of
imperial control an independence powerfully symbolized by monks
refusal to bow to rulers as a sign of respect. Not surprisingly, objections
were soon raised. The first recorded debate on monks refusal to pay
obeisance to rulers took place in 340 C.E. Another, more extensive one
occurred in 402 C.E. and it is on this one that I will focus. My aim is not
to present anything like a full account of this affair but to examine
closely some of the arguments wielded on both sides of the issue and the
language in which they were couched.26
Between 401 and 403 C.E., during an era of political division and recurrent military conflict, a provincial governor and military general,
Huan Xuan (369404), briefly usurped power and took ill-fated
steps to initiate a new dynasty. As he moved to consolidate power, one of
his first acts was to attempt to have some prominent Buddhist monks
returned to lay status, to impose a government-organized selection or
winnowing of membership in the sangha, and to assume control over
registration of monks and nuns. In this context he also chose to reopen a
great matter of our time (yidai zhi dashi ) as he called it, 27
_____________
26
This matter has been voluminously treated in Western-language scholarship: see
epecially Zrcher, Erik, The Buddhist Conquest of China, Leiden: E.J. Brill 1972, 231
239 (still the best and clearest account in a Western language); Hurvitz, Leon, Render
unto Caesar in Early Chinese Buddhism: Hui-yans Treatise on the Exemption of the
Buddhist Clergy from the Requirements of Civil Etiquette, in: Kshitis Roy (ed.), Liebenthal Festschrift, Santiniketan: Santiniketan Press 1957, 80114; and Tsukamoto, Zenry,
A History of Early Chinese Buddhism: From its Introduction to the Death of Hui-yan, Tr.
Leon Hurvitz. 2 vol. continuously paginated, Tokyo: Kodansha 1985, 828844. For general discussions of sangha-state relations in China see also Storch, Tanya, The Past Explains the Present: State Control over Religious Communities in Medieval China, in: The
Medieval History Journal 3 (2000), 311335; Yu, Anthony C., State and Religion in
China: Historical and Textual Perspectives, Chicago: Open Court 2005, 90134;
Weinstein, Stanley, Buddhism Under the Tang, Cambridge: Cambridge University Press
(period-specific but detailed coverage) 1987.
27
T 2102, 80b24.

14

ROBERT FORD CAMPANY

by calling for the carrying out of Yu Bings former plan (of 340 C.E.) to
require monks to bow before the ruler. This disturbed some of his own
followers, most notably an official named Wang Mi (360407), as
well as monastic leaders, most notably the monk Huiyuan (334
417). Many documents exchanged in the ensuing debate, all of them
written in the spring of 402, have survived. The arguments are complex
and rather voluminous. I will focus here only on those most relevant to
our topic.
Huan Xuans most basic argument for monks obeisance to rulers is
that the ruler gives life (sheng ), or at least helps sustain, regulate,
circulate, and distribute the stuff that makes life possible. Appealing to
chapter 25 of the received text of the famous Daode jing , which
numbers the king among the four greats (si da ) which also include the Dao, heaven, and earth,28 Huan argues that since monks and
nuns receive life and depend on it to exist in this world and carry out
their practices, they must be required to show respect to the one who is
responsible for managing and sustaining the life granted by heaven and
earth: the ruler.29 They may not benefit from life and the conditions of
life without paying obeisance to the one who maintains these.
Another, related argument of Huans that is of great interest for our
purposes unfolds in the following terms: Buddhism above all values
things of the spirit, or perhaps we might translate that which is spiritual (shen ). But qualities of the spirit, e.g., understanding or the lack
thereof, are possessed by individuals according to their allotted shares
(fen ). All a Buddhist teacher can do is work with his disciples natural
allotment of talents and endowments, just as the craftsman can only work
with the raw material that nature provides: he cannot create it but can
only modify it once given. The deepest virtue [in this situation] lies with
the initial provision of the material ; the subsequent
work of polishing and beautifying it is truly only secondary to this
. The ruler, not the monk, is the one involved in the
provision and maintenance of life and its conditions, and life is the
source of each individuals allotted shares; hence the way of the
_____________
28
See Mair, Victor, Tao Te Ching: The Classic Book of Integrity and the Way, New
York: Bantam 1990, 90. Some early versions of the Daodejing replace king with life;
the rude insertion of the king into this series clearly troubled some: see Boltz, William G.,
The Religious and Philosophical Significance of the Hsiang Erh Lao tzu in the Light of
the Ma-wang-tui Silk Manuscripts, in: Bulletin of the School of Oriental and African
Studies 45 (1982), 95117.113114.
29
This argument is most clearly enunciated at T 2102, 80b.

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

15

ruler encompasses , rather than being encompassed by, the

way of the teacher 30 all of which amounts to saying that the
spiritual work of Buddhist monks presupposes the prior world-ordering,
life-sustaining work of rulers and polities. This priority of origin and
function entails an obligation on the part of the sangha to acknowledge
ritually its dependence on the imperium by bowing to the ruler.
The monk Huiyuan from his monastic retreat on Mount Lu crafted a
response to these arguments.31 His treatise contains a preface that quotes
from Huan Xuans original letter on the subject, followed by five sections; the fifth, on the non-perishing of the soul at death, while fascinating, does not concern us here.
The first section treats lay Buddhists who remain in the household.
Such persons are people who acquiesce to transformation ,
where transformation (hua ) captures both the inevitable changes
that constitute life in this world and the moral transformations rulers seek
to inculcate in people through social institutions. Of lay people Huiyan
says that in their traces they accord with that which is within the realm
. He draws here on two tropes that would have been familiar to
his readers. One is that of traces (literally tracks or footprints):
people who involve themselves in ordinary life leave traces, whereas
hermits, those who withdraw from political affairs, and others who disengage themselves are said to cut off their traces or leave no traces.
The other is that of being within the realm, an expression traceable to
the inner chapters of Zhuangzi (ca. 320 B.C.E.): those who operate
within the realm of custom or etiquette abide by social convention,
whereas there are others who are beyond the realm and whose behavior cannot be judged by conventional standards or at least that is what
users of this language want their readers to believe.32 Huiyuan asserts in
this section that Buddhist householders, properly acquiescent to transformation and leaving traces in the social world, are obliged to bow before their rulers. This provision, he says, is how the teaching of Buddha
honors the provision of life and assists kingly transformation in the way
of governance .
The second section treats those who have left the household Buddhist monks and nuns. These men and women are guests beyond the
_____________
30

Text at T 2102, 83b; cf. the summary in Zrcher 1972, 234.

Text at T 2102, 29c32b. This text is introduced and translated in Hurvitz 1957; the
translation is a bit outdated and in need of improvement but suffices to give a sense of the
arguments.
32
For a brief discussion of this topos see Campany 2005, 187.
31

16

ROBERT FORD CAMPANY

realm whose traces are cut off from among [other] beings
. Because of the teaching they live by, they understand that
woes and entanglements come from having a body [or a self] and that by
not maintaining the body [or the self] one extinguishes woe
. They further know that the generation of successive lives comes from undergoing transformation and that
it is by not acquiescing to transformation that one seeks the ultimate
principle ; the ultimate principle here
connotes nirvana, the cessation of change and of rebirth (as is stated
explicitly in the next section). If extinguishing woe does not depend on
maintaining the body/self, then [monks and nuns] do not honor the benefits that foster life . Later in this
section Huiyuan speaks of monks and nuns as having crossed to beyond
transformation in order to seek the ultimate principle .
So: the ruler governs the realm of change and the processes of transformation; monks and nuns seek to move beyond these, so they are not
obliged to bow to the ruler, even though their practice benefits other
beings and assists the ruler in his work of transformation.
The third section, titled Those who seek the ultimate principle do not
acquiesce in transformation, works from the dyads established earlier to
extend the argument. It opens with a hypothetical interlocutors objection
on the grounds that understanding the ultimate principle must involve
embodying the ultimate and that such embodiment must in turn
require acquiescence in transformation. (A possible paraphrase of the
objection would be: practitioners of your teachings must exist as embodied beings; to the extent that they remain embodied beings they participate in this realm of change, and so they are the rulers subjects and must
pay him obeisance.) Huiyuan responds with another dyad: there are beings that have feelings concerning transformation and others
that do not. If one has no feelings concerning transformation, then when
transformation ends ones lives cease , whereas
if one has feelings concerning transformation one reacts and responds to
it, causing ongoing life (i.e. rebirth). Nirvana is the cessation of transformation. So it is by not encumbering ones life with feelings that
ones lives can be extinguished , and if one
does not encumber ones spirit with [repeated] lives, then ones spirit can
be made ethereal. An ethereal spirit breaking the bounds: this is what is
called nirvana .
(The term I here translate as ethereal, ming , was often used to de-

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

17

note the dark or unseen or subtle world of spirits, ghosts, demons,

etc., as opposed to the quotidian daylight world.)
The fourth section, which takes up the relationship between Buddhist
and indigenous teachings and writings, is less directly relevant for our
purposes except insofar as it continues the conceit of distinguishing between matters within the realm and those beyond the realm and links
the latter with things beyond the six directions and beyond the senses.
Like a great many arguments in traditional China and elsewhere, then,
this one proceeds by setting up an interlinked series of distinctions that
function as hierarchical dyads, one side clearly marked as superior to the
other:
+
cutting off traces
beyond the realm
not acquiescing to transformation
not involving oneself with life
being without feelings
not maintaining the body/self
beyond change

leaving traces
within the realm
acquiescing to transformation
involving oneself with
life
having feelings
having a body/self
within change

In this instance Huiyuans arguments carried the day; Huan Xuan relented, and monks and nuns retained the prerogative of not bowing before rulers. But this victory would prove short-lived, as will be seen below.
Let us pause to reflect on the terms in which Huiyuan made his case.
In positing a state or realm beyond change, beyond rulers sphere of
transformation, and beyond the realm of everyday life, a realm further
characterized (as Huan Xuan himself also said) by its attention to matters
of spirit, by being ethereal, and by having as its goal something described as ultimately authoritative and as sharply contrasted with the
concerns of ordinary human life as possible, Huiyuan came rather close
to positing a distinct arena of effort and concern that functions analogously to what would much later come to be designated, beginning in
Western European societies, as religion (as opposed to non-religion) or
the religious (as opposed to the non-religious). The analogy seems
even stronger when we consider that Huiyuan had in mind a particular
institution an institution constituted by a human tradition of teaching

18

ROBERT FORD CAMPANY

and practice as the locus in this world where the cutting off of traces
and the pursuit of the ultimate principle were carried out, in contrast to
the pursuit of life that constituted the default mode of social existence.
And, most arrestingly of all, we find this articulation of a distinct sphere
precisely at a moment when this institutions authority relative to the
imperium was being challenged. A ruler was pushing for ritual obeisance
from monks and nuns, whose growing power, status, and wealth he
feared; the argument as framed by both sides quickly became a matter of
which realm or zone that of life or that of its renunciation, that of
within the realm or that of beyond the realm was to be recognized
as having higher status. For the moment, at least, that argument won
which, having posited a distinct realm beyond this one of life and death
and change, staked its prestige on that realm and on the extraordinariness
and difficulty of reaching it successfully. The parallels to medieval
European debates on temporal versus spiritual power are striking,
though I lack space here to explore them further. It is those debates in
medieval and early modern Europe that, having carved out the religious or the spiritual as a zone of power distinct from the secular,
set the stage for the eventual rise of the generic concept of religion. It
was, I am suggesting, because the Roman Catholic Church successfully
claimed a distinct realm of power for itself not subject to the authority of
kings, and was able to back up that claim with significant social status
and financial and legal power, that religion as a domain of concern distinct from secular life religion as a sui generis category of phenomena
became thinkable, and indeed came to seem so natural that modern
scholars registered surprise when they found that many other premodern
societies lacked such a category.
But in China, despite the force and temporary success of Huiyuans
argument, the sangha was soon absorbed by the imperial state and was
never again even if it was in Huiyuans own day, which remains unclear a truly autonomous body. Monasteries could own land, but such
lands were often seized during government crackdowns on monastic
wealth and status. In theory the sangha was headed by a monk, but in
many eras that monk was appointed by the imperial court and in any case
he functioned as an official of the empire, reporting to the central court;
and from the Sui dynasty onward even this small degree of autonomy
was removed, the office in charge of the sangha being subsumed as one
of three bureaus of the Court for State Ceremonials. It was also from the
Sui on that government overseers (jian ) were placed in every Buddhist monastery and Daoist temple. By the mid-Tang, control of the

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

19

sangha was solely and powerfully in the hands of the central government, and there it would remain down to the present. And meanwhile,
looking at the other side of the coin, the Chinese empire was never, from
its inception down to its end in 1911 C.E., conceived as anything other
than a deeply (in our terms) religious system in purpose and in function. Its primary mandate was nothing less than the maintenance of
proper relations, primarily through the ritual idiom of sacrifice, between
humanity and the forces and divinities of the cosmos. The emperor was a
sacred figure; the very term we often render as emperor really meant
thearch. So the assumption that the empire should control what we
would term religious affairs was an old, indigenous assumption in
China, and the Buddhist sangha never attained enough power and status
to hold for long to any real autonomy. The already-religious nature of the
Chinese state never left the sangha (or the Daoist priesthood) much of a
separate ground on which to stand.
The shifting nomenclature of the offices charged with oversight of
monasteries and temples is of interest for our purposes, as is the shifting
placement of these offices in the larger bureaucratic structure of the empire. The details are far too numerous to treat in any detail here, but a
few examples will suffice.33 The term xuan was often used, roughly in
the sense of things beyond ordinary ken, things mysterious, a usage
dating back at least to the famous first chapter of the received text of the
Daodejing: Mystery of mysteries, the gate of all wonders! (Mair
1990:59). Under the Northern Wei dynasty, for example, the sangha was
nominally headed by a single, state-appointed monk; at various times the
bureau housing this figure was titled the Office for the Illumination of
Mysteries (zhaoxuan cao ). The Northern Qi dynasty kept this
same nomenclature but raised the office in question to the status of a
Court (si ). Under the Sui this court was downgraded again to the level
of a bureau and renamed Bureau for the Veneration of Mysteries
(chongxuan shu ). The Tang seems to have preferred to reserve
xuan for the titles of Daoist-related offices; special schools for Daoist
studies, called Chongxuan xue or Schools for the Veneration of
Mysteries, were established within the state university to prepare candi_____________
33

For a numbingly exhaustive treatment, see Yamazaki Hiroshi , Shina chsei

Bukky no tenkai , Tokyo 1942, 473516; works by Weinstein,
Chen, Barrett, the Cambridge History, etc., are also useful, although all of these works
share the unfortunate habit of examining official treatment of only one religion (Daoism or
Buddhism) at a time rather than studying them side by side, making it hard to form a
complete picture a task for future research.

20

ROBERT FORD CAMPANY

dates for an exam on the Daoist canonical scriptures modeled on civil

service exams. The old, vaguely metaphysical sense of xuan, then, suggested that what was managed under this bureaucratic rubric was somehow importantly different in kind from ordinary government affairs.
But the placement of these offices in the larger bureaucratic system
suggested otherwise. Pro-Buddhist (or at least non-anti-Buddhist) rulers
tended to subsume them under the Board of Rites, implying that the
Buddhist sangha and all that went on there was just another function
and a fully integrated function of the state, parallel to specifically Confucian rites and to the many kinds of state cults venerating everything
from gods of localities and the dynastic ancestors to Heaven and Earth.
During the reigns of anti-Buddhist rulers they tended to be placed under
the Court for State Ceremonials (honglu si ), the body responsible
for receiving foreign emissaries and conducting diplomatic relations,
implying that the sangha was seen as not fully integrated and was to be
treated as if it were a foreign body. In some cases the location of these
offices flip-flopped in the span of a year or two, as rulers changed and
ideological winds shifted at court.
Starting in the mid-Tang (the earliest mention dates to 774 C.E.), a
new office with a new name was created to oversee the Buddhist sangha:
the Commissioner of Meritorious Works (gongde shi ), responsible for merit-making activities such as erecting images, building temples
and bridges, and holding vegetarian merit feasts.34 In 807 Daoist priests
and nuns were also placed under this same jurisdiction. To the best of my
knowledge, this was the first time in Chinese history that the monastic/
priestly institutions of Buddhism and Daoism were placed on a par with
one another under the aegis of a common bureaucratic entity. We might
be tempted to see here the germ, at least, of the idea of a common genus,
of which the things we term Buddhism and Daoism were both finally being seen as species. Perhaps so. But, unlike religion(s), the
genus was not a thing separate from ordinary life or the polity: it was an
arm of imperial government. And what these two isms were being
implied to have in common was that they were both systems for the generation of the abstract, moral-theological-cosmological commodity
known as merit. Merit was, in turn, the concern not just of private
individuals or even families or local communities but also of the state as
_____________
34
For a seminal discussion of the economy of Buddhist merit-making activities, focusing on the practice of bridge-building, see Kieschnick, John, The Impact of Buddhism
on Chinese Material Culture, Princeton: 2003, 157219.

CHINESE HISTORY AND WRITING ABOUT RELIGION(S)

21

such. Not only was it a concern of the state: it was a proper function of
the state, and so the state oversaw it.
I suggest that it was largely due to this aspect of Chinese history that
no generic concept or category parallel to our religion (in its generic
sense) was developed very far. If Huiyuan had conceded that other sorts
of practitioners besides Buddhist monks and nuns might be in quest of
the ultimate principle and thus stand outside the realm or beyond
the bounds, if he had imagined his Daoist colleagues on or near Mount
Lu to be engaged on a parallel if distinct path, if he had spoken of them
too as men and women who were cutting off their traces and not acquiescing in transformation, perhaps he might have been the Chinese
progenitor of a category in Chinese history that was closely analogous to
religion as opposed to non-religion. But, if he thought in this way,
which is highly unlikely, he left no record of it. And after him, there
were few or no proposals for a separate sphere of power and activity in
the human world, corresponding to a separate level of cosmological or
theological reality, that was tolerated by any Chinese dynasty for any
length of time.35 The state itself was in our terms among other things
a religious institution; it never was thought of as inhabiting a distinct,
delimitedly secular realm of power, or rather, there was no fundamental distinction between two realms of power and concern, one of them
corresponding to what we call religious and the other to what we call
secular. The state never ceded these functions to any institutions distinct from itself. Instead, right down to 1911 and beyond, what we would
term religious institutions were almost always closely and indeed in
many cases very intensely overseen by the state when tolerated at all
and were conceptualized as branches of government that existed, in part,
to serve the state by enhancing its merit and by invoking divine or karmic blessings on it. Without a socially and institutionally distinct space
which it might have been created to name, no analogue to the Western
generic category religion ever emerged indigenously. It was, instead,
imported from Western languages by way of Japanese. Now that it has
become thoroughly rooted in Chinese discourse, it will remain there,
having forever changed the ideological, social, legal, and as we and
they would both now say religious landscape of Chinese culture.

_____________

35
One thing that there were were periodic debates at court between representatives of
Daoism and of Buddhism. One way in which I hope to expand and improve this paper is
to study closely the nomenclature and terminology used in records of these debates to see
if anything is mentioned that comes close to constituting a genus of which these two great
traditions were conceived as species. But this would be a very surprising finding.